This proposal defines and studies clustered wireless sensor networks (WSNs), WSNs whose topology due to the deployment procedure and the application requirements results in the phenomenon of clustering or clumping of nodes. It analyzes a range of topologies of clustered WSNs and their impact on the primary sensing objectives of coverage and connectivity. By exploiting the inherent advantages of clustered topologies of nodes, this proposal presents techniques for optimizing the primary performance metrics of interest in WSNs: power consumption, capacity and data transfer latency between nodes and sink in a simulated environment.

This proposal addresses the phenomenon of ‘naturally clustered’ networks of wireless sensor nodes, as opposed to WSNs where clustering is facilitated by selection. Clustering is a natural outcome of the many random deployment strategies such as spraying or scattering of nodes. In particular, the attributes of coverage in naturally clustered networks are extensions of the Poisson model of node distribution widely used in modeling the distribution of nodes in WSNs. We show how to extend Poisson distribution of nodes to support clustering by examining how to create an analytical framework for naturally clustered networks using Poisson cluster point process (PCPP) to simulate the phenomenon of clustering in the placement of nodes. The knowledge of coverage is essential in developing algorithms for optimizing the tradeoff between coverage and power consumption of dense clustered WSNs. We show that along with increasing the vacancy in random placement of nodes in a WSN, it also alters the connectivity properties in the network. We analyze clustering in the presence of obstacles, and study varying levels of redundancy to determine the probability of coverage in the network. Our proposed models for clustered WSNs embrace the domain of a wide range of topologies that are prevalent in actual real-world deployment scenarios, and call for clustering-specific protocols to enhance network performance. To address this issue, we develop power management algorithms tailored to various clustering scenarios with an aim to optimize the level of active coverage and maximize the network lifetime.

Committee Members:

Dr. Sirin Tekinay (Advisor), Associate Professor, Department of ECE, NJIT
Dr. Guiling Wang (Co-advisor), Assistant Professor, Department of Computer Science, NJIT
Dr. Nirwan Ansari (Committee Chair), Professor, ECE Department, NJIT
Dr. Roberto Rojas-Cessa, Associate Professor, Department of ECE, NJIT
Dr. Yanchao Zhang, Assistant Professor, Department of ECE, NJIT
Dr. Tiffany Jing Li, Associate Professor, Department of ECE, Lehigh University

Click here for seminar archive

Note: All MS thesis and PhD dissertation (proposal) defense are counted towards ECE791.